Non-projective Parsing for Statistical Machine Translation
Xavier Carreras and Michael Collins
In: EMNLP 2009, 6-7 August 2009, Singapore.
We describe a novel approach for syntax-based statistical MT, which builds on a variant of tree adjoining grammar (TAG). Inspired by work in discriminative dependency parsing, the key idea in our approach is to allow highly flexible reordering operations during parsing, in combination with a discriminative model that can condition on rich features of the sourcelanguage string. Experiments on translation from German to English show improvements
over phrase-based systems, both in terms of BLEU scores and in human evaluations.